Bug 2210287
| Summary: | The "dev" setting is ignored when it is used without address for passt interface | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 9 | Reporter: | yalzhang <yalzhang> |
| Component: | libvirt | Assignee: | Michal Privoznik <mprivozn> |
| libvirt sub component: | Networking | QA Contact: | yalzhang <yalzhang> |
| Status: | CLOSED ERRATA | Docs Contact: | |
| Severity: | unspecified | ||
| Priority: | unspecified | CC: | dgibson, dzheng, haizhao, jdenemar, lmen, lvivier, mprivozn, sbrivio, virt-maint |
| Version: | 9.3 | Keywords: | AutomationBackLog, Triaged, Upstream |
| Target Milestone: | rc | Flags: | pm-rhel:
mirror+
|
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | libvirt-9.5.0-0rc1.1.el9 | Doc Type: | If docs needed, set a value |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2023-11-07 08:31:41 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | 9.5.0 |
| Embargoed: | |||
Patch posted on the list: https://listman.redhat.com/archives/libvir-list/2023-June/240148.html To POST:
commit 1c7335add9e28637d8a8b5039f487e5dc6a591c2
Author: Michal Prívozník <mprivozn>
AuthorDate: Thu Jun 1 10:37:04 2023 +0200
Commit: Michal Prívozník <mprivozn>
CommitDate: Thu Jun 1 14:25:08 2023 +0200
qemu_passt: Format portForward device even without address
It's almost like we've anticipated this. Our XML parser and
formatter handles @address and @dev attributes of <portForward/>
element completely independent of each other. And as of commit
2023_03_29.b10b983~3 passt allows handling these two separately
too. All that's left is generate the cmd line according to this
new fact.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=2210287
Signed-off-by: Michal Privoznik <mprivozn>
Reviewed-by: Ján Tomko <jtomko>
v9.4.0-3-g1c7335add9
Hi Michal, the vm can not start successfully(See S3 below). Maybe it depends on a new passt build. Could you please help to confirm? Thank you!
S1: with both dev and address:
$ virsh dumpxml test1 --xpath //interface
<interface type="user">
<mac address="52:54:00:26:44:e5"/>
<source dev="eno1"/>
<portForward proto="tcp" dev="eno1" address='10.73.178.71'>
<range start="41335"/>
</portForward>
<model type="virtio"/>
<backend type="passt"/>
<address type="pci" domain="0x0000" bus="0x01" slot="0x00" function="0x0"/>
</interface>
will come up with:
$ ps aux | grep '/usr/bin/passt'
test 13544 0.0 0.2 76344 38124 ? Ss 01:20 0:00 /usr/bin/passt --one-off --socket /run/user/1000/libvirt/qemu/run/passt/2-test1-net0.socket --mac-addr 52:54:00:26:44:e5 --pid /run/user/1000/libvirt/qemu/run/passt/2-test1-net0-passt.pid --interface eno1 --tcp-ports 10.73.178.71%eno1/41335
S2: only address
<portForward proto="tcp" address='10.73.178.71'>
<range start="41335"/>
</portForward>
VM start successfully and the passt process is like this:
$ ps aux | grep '/usr/bin/passt'
test 13622 0.0 0.2 76344 38828 ? Ss 01:21 0:00 /usr/bin/passt --one-off --socket /run/user/1000/libvirt/qemu/run/passt/3-test1-net0.socket --mac-addr 52:54:00:26:44:e5 --pid /run/user/1000/libvirt/qemu/run/passt/3-test1-net0-passt.pid --interface eno1 --tcp-ports 10.73.178.71/41335
S3: only dev, vm can not start successfully
<portForward proto="tcp" dev="eno1">
<range start="41335"/>
</portForward>
$ virsh start test1
error: Failed to start domain 'test1'
error: internal error: Child process (/usr/bin/passt --one-off --socket /run/user/1000/libvirt/qemu/run/passt/4-test1-net0.socket --mac-addr 52:54:00:26:44:e5 --pid /run/user/1000/libvirt/qemu/run/passt/4-test1-net0-passt.pid --interface eno1 --tcp-ports %eno1/41335) unexpected exit status 1: Invalid port specifier %eno1/41335
This BZ (it's pre-verification actually) is blocking libvirt packages from making it into a RHEL compose. Could this be pre-verified earlier than after the passt rebase in mid August? For example using passt from upstream or using a scratch build (aka rebase pre-release) or something similar? Test with upstream passt and libvirt, the result is as expected: # rpm -q passt libvirt passt-g023d684-1.x86_64 libvirt-9.5.0-3.el9.x86_64 # getenforce Permissive 1) with address and interface: # virsh dumpxml rhel --xpath //portForward <portForward proto="tcp" address="10.73.178.37" dev="eno1"> <range start="41335"/> </portForward> the passt cmdline: # ps aux | grep 'usr/bin/passt' | grep -v grep qemu 5574 0.0 0.0 72936 37628 ? Ss 02:54 0:00 /usr/bin/passt --one-off --socket /run/libvirt/qemu/passt/7-rhel-net0.socket --mac-addr 52:54:00:49:84:12 --pid /run/libvirt/qemu/passt/7-rhel-net0-passt.pid --interface eno1 --tcp-ports 10.73.178.37%eno1/41335 2) with address only: # virsh dumpxml rhel --xpath //portForward <portForward proto="tcp" address="10.73.178.37"> <range start="41335"/> </portForward> the passt cmd line: # ps aux | grep 'usr/bin/passt' | grep -v grep qemu 5692 0.0 0.0 72932 36628 ? Ss 02:55 0:00 /usr/bin/passt --one-off --socket /run/libvirt/qemu/passt/8-rhel-net0.socket --mac-addr 52:54:00:49:84:12 --pid /run/libvirt/qemu/passt/8-rhel-net0-passt.pid --interface eno1 --tcp-ports 10.73.178.37/41335 3) with interface only: # virsh dumpxml rhel --xpath //portForward <portForward proto="tcp" dev="eno1"> <range start="41335"/> </portForward> the passt cmdline: # ps aux | grep 'usr/bin/passt' | grep -v grep qemu 5779 0.0 0.0 72932 36632 ? Ss 02:56 0:00 /usr/bin/passt --one-off --socket /run/libvirt/qemu/passt/9-rhel-net0.socket --mac-addr 52:54:00:49:84:12 --pid /run/libvirt/qemu/passt/9-rhel-net0-passt.pid --interface eno1 --tcp-ports %eno1/41335 Test with lastest passt downstream build, the result is as expected. libvirt-9.5.0-5.el9.x86_64 passt-0^20230818.g0af928e-4.el9.x86_64 passt-selinux-0^20230818.g0af928e-4.el9.noarch Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Moderate: libvirt security, bug fix, and enhancement update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2023:6409 |
Description of problem: The "dev" setting is ignored unexpectedly when it is used without address for passt interface Version-Release number of selected component (if applicable): libvirt-9.3.0-2.el9.x86_64 How reproducible: 100% Steps to Reproduce: 1. Prepare a vm of user type interface with passt backend as below: # virsh dumpxml rhel --xpath //interface <interface type="user"> <mac address="52:54:00:49:84:12"/> <source dev="eno1"/> <portForward proto="tcp" dev="eno1"> <range start="41335"/> </portForward> <model type="virtio"/> <backend type="passt"/> <address type="pci" domain="0x0000" bus="0x01" slot="0x00" function="0x0"/> </interface> 2. Start vm and check the passt process # ps aux | grep 'bin/passt' | grep -v grep qemu 9303 0.0 0.0 76344 33524 ? Ss 08:59 0:00 /usr/bin/passt --one-off --socket /run/libvirt/qemu/passt/7-rhel-net0.socket --mac-addr 52:54:00:49:84:12 --pid /run/libvirt/qemu/passt/7-rhel-net0-passt.pid --interface eno1 --tcp-ports 41335 The "dev" setting is ignored unexpectedly, it should be like "--tcp-ports %eno1/41335" Actual results: The "dev" setting is ignored unexpectedly when it is used without address Expected results: The "dev" setting should be effective in the cmd line "--tcp-ports %eno1/41335" Additional info: And suggest to mention the "dev" attribute in the docs/formatdomain.rst